Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Mar 15.
Published in final edited form as: Stem Cells. 2015 Jun 23;33(9):2738–2747. doi: 10.1002/stem.2068

p38γ MAPK is a therapeutic target for triple-negative breast cancer by stimulation of cancer stem-like cell expansion

Xiaomei Qi 1, Ning Yin 1, Shao Ma 1, Adrienne Lepp 1, Jun Tang 2, Weiqing Jing 3, Bryon Johnson 3, Michael B Dwinell 4, Christopher R Chitambar 3, Guan Chen 1,5
PMCID: PMC4792188  NIHMSID: NIHMS756948  PMID: 26077647

Abstract

Triple-negative breast cancer (TNBC) is highly progressive and lacks established therapeutic targets. p38γ mitogen-activated protein kinase (MAPK) (gene name: MAPK12) is overexpressed in TNBC but how overexpressed p38γ contributes to TNBC remains unknown. Here we show that p38γ activation promotes TNBC development and progression by stimulating cancer stem-like cell (CSC) expansion and may serve as a novel therapeutic target. p38γ silencing in TNBC cells reduces mammosphere formation and decreases expression levels of CSC drivers including Nanog, Oct3/4, and Sox2. Moreover, p38γ MAPK-forced expression alone is sufficient to stimulate CSC expansion and to induce epithelial cell transformation in vitro and in vivo. Furthermore, p38γ depends on its activity to stimulate CSC expansion and breast cancer progression, indicating a therapeutic opportunity by application of its pharmacological inhibitor. Indeed, the non-toxic p38γ specific pharmacological inhibitor pirfenidone selectively inhibits TNBC growth in vitro and/or in vivo and significantly decreases the CSC population. Mechanistically, p38γ stimulates Nanog transcription through c-Jun/AP-1 via a multi-protein complex formation. These results together demonstrate that p38γ can drive TNBC development and progression and may be a novel therapeutic target for TNBC by stimulating CSC expansion. Inhibiting p38γ activity with pirfenidone may be a novel strategy for the treatment of TNBC.

Keywords: p38γ MAPK, triple negative breast cancer, therapeutic target, cancer stem-like cells, c-Jun/AP-1

Introduction

Cancer stem-like cells (CSCs) represent a sub-population of cells with high self-renewal capacity and possible tumor-initiating properties (1). The CSC population is exchangeable with the non-CSC counterpart and is driven by a group of master transcription factors including Nanog, Oct3/4 and Sox2 (2). These factors are typically active in embryonic stem cells, silenced in adult cells, but re-activated again in cancer (2). Recent studies showed that CSC may be responsible both for cancer initiation and progression, suggesting that targeting CSC may be an important anti-cancer strategy (3). However, upstream activators and pathways that control these transcription factors are mostly unknown (2). Identification of novel druggable CSC activators will be fundamentally important for cancer prevention and treatment by targeting CSC (4).

Triple-negative breast cancer (TNBC) lacks estrogen receptor (ER), progesterone receptor (PR), and epidermal growth factor receptor 2 (Her-2) for therapeutic targeting and has the worst prognosis among all types of breast cancer (5, 6). TNBC accounts for approximately 15% of breast cancer cases in women (5). The majority of TNBC share the gene expression profiles and histological characterization with basal-like breast cancer (BBC), with both displaying an aggressive clinical behavior and frequently occurring in younger and premenopausal patients (5). TNBC is highly heterogeneous but is enriched with the CSC population (7). These results indicate that CSC may be involved in TNBC development and progression, and targeting CSC may be a novel therapeutic strategy against TNBC.

p38γ is a oncogenic mitogen-activated protein kinase (MAPK) (gene name: MAPK12) and is activated through both overexpression (8) and phosphorylation (9). p38γ stimulates breast cancer invasion through collaboration with Ras to antagonize ER activity (10, 11). Recent studies from several laboratories further demonstrated that p38γ is an important breast cancer metastasis gene and is specifically overexpressed in TNBC (11-14). Moreover, p38γ overexpression is associated with decreased survival of breast cancer patients (13). These results together indicate that p38γ may contribute to TNBC development and progression. However, how p38γ contributes to TNBC remains mostly unknown. Here we tested the hypothesis that p38γ may be a novel TNBC driver by activating the CSC program. Our results show that p38γ actively stimulates CSC expansion by trans-activating Nanog via c-Jun/AP-1, which is involved in TNBC transformation and progression. Most importantly, the non-toxic p38γ specific pharmacological inhibitor pirfenidone (PFD) inhibits the CSC program and selectively blocks TNBC growth in vitro and/or in vivo. These results together indicate that p38γ may promote TNBC development and progression by stimulating CSC expansion and targeting p38γ by PFD may be a novel therapeutic strategy for the treatment of TNBC.

Materials and Methods

Cell culture, constructs, and reagents

MCF10A, 293T, and all other human breast cancer cells were purchased from ATCC (Manassas, VA). Breast cancer cells are cultured in MEM medium with 10% serum and antibiotics (11). MCF10A/Vector and MCF10A/p38γ cells are cultured in DMEM/F12 medium containing 10% heat-inactivated horse serum, EGF (20 ng/ml), hydrocortisone (0.5 μg/ml), cholera toxin (100 ng/ml) and insulin (5 μg/ml). Flag-tagged p38γ in pCDNA3 and adenoviral vector was provided by Dr. Jiahuai Han (the Scripps Research Institute) (15), which was cloned into retroviral pLHCX vector in our laboratory (10). MKK6-p38γ (constitutively active, CA) and MKK6-p38γ/AGF (dominant negative, DN) were expressed in Tet-on system in MCF-7 cells (11). The human Nanog luciferase reporter and its M1 (AP-1 like) site mutated form were provided by Dr. A. S. Nateri (University of Nottingham, UK) (16). Luciferase-expressing 231 cells were stably depleted of p38γ protein by lentiviral infection and a separate antibiotic selection (17). The pLenti6/Block-iT system (Block-it™ U6 RNAi entry vector Kit and Block-it™ Lentiviral RNAi expression system, Invitrogen) was used to clone shRNA sequences. The target sequences for control luciferase (shLuc) and for p38γ (sh-#1 and #2 p38γ) are the same as previously described (17). Pirfenidone (PFD) was purchased from Sigma (St. Louis, MO), whereas 17-AAG was obtained from Selleck (Houston, TX). Both of these compounds were dissolved in DMSO as a stock solution and stored at −20°C until use. Antibody against pan Ras was purchased from Oncogene Sciences (Cambridge, MA) and against p38γ was from R&D Systems (Minneapolis, MN). Phospho-specific antibodies were from Cell Signaling (Beverly, MA), while all others were from Santa Cruz Biotechnology (Santa Cruz, CA). Cell culture reagents were from Gibco (Grand Island, NY) and other chemicals were from Sigma.

Mammosphere formation, soft-agar, colony formation, cell invasion, cell viability, and in vitro transformation assays

Tumor mammosphere formation was performed as previously described with minor modifications (18). For breast cancer cells, we used MEM medium containing 5 μg/ml insulin, 5 μg/ml hydrocortisone, 20 ng/ml EGF, and 0.5% FBS. For MCF10A cells, cells were plated for sphere formation in medium containing 0.5% of horse serum. To assess sphere formation, cells (2.5 and 5 × 104 cells/well/ 2 ml for breast cancer and immortal mammary epithelial cells, respectively) were cultured as single-cell suspension in ultralow attachment plate (BioExpress and Costar) for about 10 days, with 0.2 ml of fresh medium added every 2-3 days. Mammospheres formed were then photographed and counted. Colony formation assays were conducted as previously described (9). For cell invasion, 2 × 105 cells were seeded in a Matrigel invasion-chamber (BD Biosciences) and assessed for their invasive activity according to the Manufacturer’s instructions (10). For viability assay, cells were cultured with 2 μM of 17-AAG or solvent for 24 h. Thereafter, cell viability was determined by trypan blue exclusion assays (9). For in vitro transformation assays, cells at early passages were plated in growth media containing 0.33% Sea-plaque-agarose (8). Two weeks later, colonies formed were visualized by staining, which were photographed and counted (17).

Animal studies

Experiments in nude mice were approved by the Medical College of Wisconsin Institutional Animal Care and Use Committee. Female Balb/c nude mice (5-6 week old) were purchased from Harlan. To assess tumor-forming ability in vivo, MCF10A/p38γ and MCF10A/Vector cells at early passage were suspended in 50% (V/V) Matrigel and 5 × 06 of these cells in 100 μl were s.c. injected into both flanks of nude mice. For the in vivo tumorigenesis assay, 231 cells (with and without stable p38γ knockdown in vitro) at different concentrations in 100 μl of 50% Matrigel were s.c. injected into female nude mice. Tumorigenesis and tumor-growth were monitored and recorded (19). To examine metastasis, MDA-MD-231 (231) cells were first modified to stably express the Firefly luciferase gene. The cells were then depleted of p38γ by lentiviral infection. Resultant cells were suspended in 100 μl of PBS (5 × 105) and were i.v. injected vial lateral tail veins of mice. Mice were monitored for tumor metastasis using a Lumina IVIS-100 In Vivo Imaging System (Xenogen Corpt, Alameda, CA, USA). For PFD anti-tumor studies, cells in 100 μl of cold PBS (2 × 106) were s.c. injected into both flanks of nude mice. When tumors became palpable, mice were randomly divided into two groups. PFD solution and solvent DMSO were i.p. administrated as described in figure legends. Tumor volumes were measured and calculated as described (17).

Statistical analysis

Results were compared using student’s t test, unless otherwise indicated. P values less than 0.05 were considered significant.

Results

p38γ is required for the maintenance of CSC population in TNBC

Recent studies show that CSC population is enriched in TNBC (7, 20). For example, more than 85% of TNBC MDA-MB-231 (231) cells are CD44 positive and CD42 negative (CD44+/CD24−) (20). Because p38γ is overexpressed in TNBC (11-14), we first determined if endogenous p38γ is required for CSC maintenance. TNBC 231 and MDA-MB-468 (468) cells (6) were stably depleted of p38γ by lentiviral mediated shRNA expression (17) and their mammosphere forming activity was assessed (18). Results (Figs.1A-C) show that p38γ knockdown significantly reduces the sphere formation in both cell lines, indicating that p38γ is required for maintaining CSC population in TNBC cells. Moreover, p38γ silencing also significantly decrease RNA levels of the key CSC drivers in these cells (Figs.1D/E), including the transcription factor Nanog, Oct3/4, and Sox2 (2). A decreased expression of Nanog, Oct3/4, and CD44 by p38γ knockdown was further demonstrated at protein levels in both cell lines (Figs.1C/F/G, Sox2 and CD24 undetectable). In addition, p38γ knockdown in 231 cells also decreases the tumorigenesis and tumor-growth in vivo in association with decreased Oct3/4 and CD44 protein expression in tumor tissues (Supplementary Figs.S1A-C). Furthermore, incubation of TNBC cells with the p38γ (but not p38α or p38β) specific pharmacological inhibitor pirfenidone (PFD) (9, 11, 21, 22) also inhibits sphere formation and decreases Nanog, Oct3/4, and Sox2 expression (Supplementary Figs.S1D-F). Together, these results demonstrate that elevated p38γ MAPK in TNBC cells play an important role in maintaining CSC population.

Figure 1. p38γ is required for mammosphere formation and for Nanog, Sox2, Oct3/4 and CD44 expression in TNBC cells.

Figure 1

Open in a new tab

A-C, p38γ depleted and control cells were plated in mammosphere culture for about 10 days and mammospheres formed were photographed and quantitated. Representative mammosphere-images are given in C (left, scale bar = 100 μM) and quantitative results are presented in A and B (mean ± SD, n = 3). Effects of p38γ depletion on Nanog and Oct3/4 protein expression were analyzed by Western blot (WB) (C, right). D, E, total RNAs were prepared and levels of Nanog, Oct3/4, and Sox2 expression were assessed by quantitative RT-PCR as previously described (43). Results are relative to those infected with control shLuc after normalization to GAPDH (± SD, n = 3). F, G, protein samples were prepared from p38γ-depleted and shLuc TNBC cells and analyzed for protein expression. Representative WB results are given at left and quantitative results are at right (relative to shLuc, measured by the NIH image after normalization to the control α-Actinin or β-actin, n = 3, bars, SD).

p38γ forced-expression alone is sufficient to increase CSC population and to induce mammary epithelial cell transformation in vitro and in vivo

To investigate if p38γ forced-expression alone is sufficient to stimulate CSC expansion, which may contribute to the TNBC oncogenesis, immortal human mammary epithelial MCF10A cells were stably expressed with p38γ by retroviral infection (10). Transduced cells were then analyzed for protein expression and morphological alterations. MCF10A cells lack endogenous ER, PR, and Her-2 protein expression (Supplementary Fig.S2A) and are widely used for CSC and transformation studies (23, 24). Results (Fig.2A) show that p38γ expressing cells are spindle-like and more reflectile with a pattern of protein expression consistent with a phenotype of the epithelial-mesenchymal-transition (EMT) as compared to vector cells (25-27). Moreover, p38γ stimulates sphere formation, increases CSC drivers’ expression and levels of ectopically expressed p38γ are further elevated in mammosphere culture as compared to adherent conditions (Figs.2A-C). These results indicate that p38γ overexpression stimulates CSC expansion. Furthermore, MCF10A/p38γ cells form much more colonies on soft-agar (Fig.2B, right panel). While a stable expression of ERK2 in MCF10A cells can result in an EMT phenotype (28), our results for the first time demonstrate an ability of a MAPK to confer an anchorage-independent growth by overexpression, indicating a transforming activity of p38γ MAPK. Consistent with this notion, the s.c. inoculation of MCF10A/p38γ cells into female nude mice resulted in tumor formation in 4 of 5 injections, whereas none of 5 injections with the same number of vector cells yielded a noticeable nodule (Fig.2D). Immunohistochemistry (IHC) analyses further show that there are similar patterns of TNBC marker expression in MCF10A/p38γ tumors and cells (Fig.2A and Supplementary Fig.S2B). Importantly, p38γ-induced Nanog protein expression was demonstrated both in MCF10A/p38γ cells and tumors (Figs.2A/D), indicating its involvement in CSC expansion during the transformation process in vitro and in vivo. The p38γ overexpression in another human mammary epithelial cell line (HMEC) also induces soft-agar growth, stimulates sphere formation, and increases CSC drivers’ expression (Supplementary Figs.S2C-H). Together, these results demonstrate that a forced-expression of p38γ MAPK alone stimulates CSC expansion and malignant transformation in immortal human breast epithelial cells in vitro and/or in vivo.

Figure 2. p38γ overexpression stimulates CSC expansion, induces transformation in vitro and in vivo, and increases expression levels of CSC/TNBC markers.

Figure 2

Open in a new tab

A, MCF10A cells were stably expressed with p38γ by retroviral transduction. Resultant cells were analyzed for protein expression by WB (left, similar results obtained from a separate experiment) and for morphological alterations and mammosphere formation (right, scale bar = 100 μM for all panels). B, p38γ and vector transduced cells were assessed for mammosphere formation (left), for Nanog, Oct3/4, and Sox2 expression by qRT-PCR (middle), and for soft-agar growth (right). Results shown are means of 3-4 experiments (± SD). C, cells were cultured in normal (Adhere; adherent) or mammosphere (Mammo) conditions and collected for WB analyses. D, indicated cells (5 × 106) were suspended in 100 μl of 50% of Matrigel, which was s.c. injected to both flanks of female Balb/c nude mice for tumor formation as indicated. Tumors at the indicated time were photographed and weighed (inserts). Also, MCF/10A/p38γ tumors were processed for immunohistochemistry (IHC) for Nanog/p38γ protein expression (53).

p38γ must be phosphorylated to promote breast cancer progression and to stimulate CSC expansion in TNBC but not in non-TNBC cells

Although CSC has been considered to be a target for cancer therapy, there is a lack of specific pharmacological inhibitors to target CSC population (29). Increased invasion and metastasis are major characteristics of both TNBC (5) and CSC (30). Studies from us (10, 11) and others (12, 13) have demonstrated a stimulatory activity of p38γ on breast cancer invasion but whether p38γ phosphorylation is required for this action is unknown. This is fundamentally important for targeting p38γ-induced CSC by using its pharmacological inhibitor. A tetracycline inducible (Tet-on) system was used to express a constitutively active (CA) p38γ (the MKK6-p38γ fusion protein) and its dominant negative (DN) form (MKK6-p38γ/AGF) in ER positive MCF-7 cells (11). After an overnight incubation with and without Tet, cells were assessed for protein expression and Matrigel invasion. Results (Fig.3A) show that CA p38γ increases and its DN mutant decreases invasion, indicating an invasion-stimulating activity of p38γ depending on its phosphorylation. Although CA p38γ increases CD44/Oct3/4 and decreases CD24 levels in MCF-7 cells, it did not significantly stimulate sphere formation (Figs.3B-D). Similarly, adenoviral mediated p38γ expression stimulates invasion (10), but not sphere formation in ER+ T47D cells despite of its increases of expression levels of several CSC markers (Figs.3B-D). However, p38γ overexpression in TNBC cells can still stimulate CSC expansion and endogenous p38γ proteins are further up-regulated in mammosphere culture (Supplementary Fig.S3A). Moreover, in contrast to TNBC cells (Supplementary Figs.1D-F), incubation of Her-2 positive BT474 cells with PFD has no significantly inhibitory effect on sphere formation. This occurs even though PFD can dose-dependently inhibit the p38γ activity {a decrease in phosphorylated protein tyrosine phosphatase H1 (p-PTPH1; PTPH1 is a specific p38γ substrate) (9) (Figs.3E/F). The lack of the significant effects of p38γ expression and inhibition on sphere formation in non-TNBC cells may be due to the p38γ antagonistic activity of ER signaling (10). These results further indicate a CSC-stimulating activity of p38γ in TNBC but not in non-TNBC cells. Consistent with the in vitro data (10), knockdown of endogenous p38γ from TNBC 231 cells decreases metastasis in mice (Supplementary Figs.S3B-D). Moreover, MCF10A/p38γ cells are more invasive and resultant invasion was significantly blocked by PFD in association with a downregulation of Nanog, Sox2, and Oct3/4 (Figs.3G/H). Together, these results indicate that p38γ must be phosphorylated to stimulate invasion and CSC expansion and thereby provide evidence for targeting CSC by using its pharmacological inhibitor PFD.

Figure 3. p38γ requires its activity to increase invasion and to stimulate CSC expansion.

Figure 3

Open in a new tab

A-D, indicated cells were assessed for protein expression, the Matigel invasion, and CSC expansion 24 h after incubation with and without Tet addition (1 μg/ml, MCF-7) or 48 h after adenoviral infection (T47D, Cont.; infected with ad-β-Gal; p38γ; infected with ad-p38γ). Results (A, B, and D) are mean of 3 separate experiments (± SD), with the insert in A showing p38γ expression and phosphorylation. Please note that the mammosphere numbers of T47D cells (D) are less than those in Supplementary Fig.7B likely as a result of adenoviral infection-associated toxicities. E, F, cells were analyzed for sphere formation (E, mean ± SD, n =3) and for protein expression (F) after incubation with DMSO or PFD (25 and 50 μg/ml) for about 10 days (E) or overnight (F). G, indicated cells were seeded in a Matrigel-invasion chamber in the absence and the presence of 20 μg/ml of PFD or solvent overnight and invaded cells were stained and manually counted (± SD, n = 3). H, MCF10A/p38γ cells were incubated with PFD or solvent as in G, and then analyzed by qRT-PCR (mean ± SD, n = 3).

p38γ is required for Ras expression through a complex formation with Hsp90, which plays an important role in TNBC survival

Ras is overexpressed in up to 50% of breast cancer (31-33) and contributes to CSC expansion (34, 35) and TNBC transformation (36). Our previous studies showed that oncogenic Ras stimulates p38γ expression in several cell lines (8, 10, 17). Because levels of Ras and p38γ protein expression are both elevated in TNBC cells (11), we next examined if p38γ may regulate Ras expression, thereby contributing to CSC expansion and TNBC phenotype. Of great interest, p38γ knockdown decreases Ras protein levels, as detected with a pan Ras antibody, in TNBC cells (Fig.4A, left, and Supplementary Fig.S4A). Moreover, p38γ-forced expression by adenovirus also increases Ras protein levels, which is blocked by incubation with PFD (Supplementary Fig.S4B). These results further indicate that p38γ activity plays an important role in Ras expression in TNBC cells.

Figure 4. p38γ and Hsp90 cooperate to maintain Ras protein expression in TNBC cells.

Figure 4

Open in a new tab

A, p38γ depleted cells were analyzed for Ras protein expression by WB (left). Indicated MCF10A cells were treated with 0.5 μM of 17-AAG or solvent for 24 h, which were then subjected to p38γ or Hsp90 immunoprecipitation (IP)/WB analysis (middle), with a portion of lysates analyzed by direct WB (right). B, cells were treated with 17-AAG or solvent for 24 h, which were then subjected to WB analyses (similar results obtained by a separate WB). C, indicated cells were incubated with 17-AAG (1 μM) or DMSO for 24 h and cell viability was then determined by trypan blue exclusion assays (± SD, n = 3).

We recently showed that p38γ can maintain Ras expression in colon cancer cells through activating heat shock protein 90 (Hsp90) (22) and wished to explore if the same mechanism operates in p38γ increasing Ras expression in breast cancer. p38γ transformed and vector transduced MCF10A cells were next analyzed for protein expression after incubation with and without the Hsp90 inhibitor 17-AAG. Results (Fig.4A, right) show that there is increased level of Ras protein in MCF10A/p38γ over vector cells, which is suppressed by the Hsp90 inhibitor 17-AAG. These results indicate that p38γ may induce TNBC transformation through collaboration with Ras via the Hsp90 activity. Hsp90 is a conserved chaperone protein and protects overexpressed or mutated oncoprotein from proteasome-dependent degradation through a complex formation (37). We next explored if p38γ forms a complex with Hsp90 and Ras, and thereby maintains the overexpressed Ras. Analysis of Hsp90 and p38γ immunoprecipitates show their interaction in MCF10A/p38γ, but not in MCF10A/Vector, cells, which was again disrupted by 17-AAG (Fig.4A, right). Similar results were also obtained in TNBC 231 cells (Supplementary Fig.S4C). Consistent with our previous results (8, 10, 17), overexpression of an oncogenic H-Ras (G12V) in MCF10A cells also increases levels of p38γ, but not its isoform p38α, expression (Supplementary Fig.S4D). Together, these results indicate that p38γ and Ras stimulate each other expression in TNBC cells through Hsp90 activity.

Hsp90 inhibitors were previously shown to inhibit TNBC growth in vitro and in vivo (38) but the underlying mechanisms remain unknown. Because p38γ is overexpressed in TNBC (11-14) and can activate Hsp90 (22), we next determined if Hsp90 is required for Ras expression and cell survival for TNBC cells. Results in Fig.4B show that incubation with 17-AAG only depletes Ras protein in TNBC, but not in non-TNBC, cells, whereas it decreases levels of Raf-1, an established Hsp90 client protein (39), in every cell line. More importantly, the treatment with 17-AAG significantly induces cell-death in TNBC, but not in non-TNBC, cells (Fig.4C). Since p38γ is required for CSC maintenance and Ras expression through Hsp90 (Figs.1/2/4 and Supplementary Figs. S1/2/4), these results indicate that one of the mechanisms by which p38γ cooperates with Ras and Hsp90 in linking the CSC program to the TNBC phenotype is to promote cell survival.

p38γ may be a therapeutic target for TNBC by stimulating CSC expansion

Our results so far have demonstrated an active role of p38γ in TNBC transformation, invasion/metastasis, and survival by stimulation of CSC expansion through cooperation with Hsp90 and Ras (Figs.1-4). Since p38γ is overexpressed in TNBC (10-14), we next directly examined if p38γ is a therapeutic target for TNBC by using its specific inhibitor PFD (21). Breast cancer cells were incubated with PFD or vehicle control DMSO for about 2 weeks and resultant effects on colony formation were determined (11). Results (Fig.5A and Supplementary Figs.4E-H) showed that p-PTPH1 levels are decreased by PFD and increased by p38γ overexpression, and PFD inhibits colony formation in TNBC (but not in non-TNBC) cells. Together with the CSC-inhibitory activity of PFD as well as p38γ knockdown-induced CSC depletion (Figs.1 and Supplementary Fig.S1), these results strongly suggest that p38γ may be a therapeutic target for TNBC through maintaining CSC population.

Figure 5. Pirfenidone (PFD) selectively inhibits the colony formation of TNBC, but not non-TNBC, cells in vitro and significantly suppresses the TNBC xenograft-growth in nude mice.

Figure 5

Open in a new tab

A, indicated cells were cultured with PFD or solvent for about 2 weeks, and effects on growth were assessed by clonogenic assays (± SD, n = 3-4). B, C, indicated cells (2 × 106 in 100 μl of PBS) were s.c. injected into right and left flank areas of female Balb/c nude mice, that were then subjected to the PFD therapy at indicated time. PFD was i.p. administrated daily into mice at 250 mg/kg/day on days 9-11, and at 125 mg/kg/day on days 18-20. Control groups were injected with the same volume of DMSO. Tumor volume was measured with a caliper and results shown are the means of 4 to 5 mice (± SD). D, E, PFD treatment decreases tumor weights. By the end of experiments, tumors were excised and weighed (mean ± SD, n = 4).

PFD is an oral-active and relatively non-toxic anti-fibrotic agent (40), and has been approved by FDA for the treatment of lung fibrosis (41, 42). To further demonstrate if PFD has new therapeutic potentials in TNBC, 231 and 468 cells were injected into nude mice to form xenografts. When tumors are palpable, the therapy with PFD or control solvent was initiated. Results (Figs.5B-E) show that PFD significantly suppresses the tumor growth in both TNBC xenografts as compared to DMSO control. Consistent with the effects in vitro (Supplementary Figs.S1E/F), the systemic therapy with PFD also substantially decreases RNA levels of Nanog, Sox2 and Oct3/4 in tumor tissues (Supplementary Figs.S5A/B). Moreover, p38γ knockdown decreases colony formation in TNBC cells (Supplementary Figs.S5C-E), further supporting its promoting role in TNBC growth. To further demonstrate if PFD inhibits the TNBC tumor growth through targeting CSC population, RNAs were prepared from DMSO and PFD treated tumors, which were subjected to CSC arrays using a Qiagen Kit. Results (Supplementary Fig.S6) showed that expression levels of a group of CSC genes are mostly down-regulated by the PFD treatment. These results further demonstrate that PFD may suppress TNBC growth in vitro and in vivo by targeting CSC population. Consistent with its lack of inhibitory effect on colony formation (Fig.5A), PFD also failed to suppress sphere-formation in MCF-7 and T47D cells (Supplementary Figs.7A/B). Together, these results indicate that targeting p38γ by its specific pharmacological inhibitor PFD can selectively suppress TNBC growth through decreasing CSC population in vitro and/or in vivo.

p38γ trans-activates Nanog through the c-Jun/AP-1-mediated promoter binding

Nanog, Oct4, and Sox2 constitute the regulatory core circuitry of key transcription factors in driving stem cell initiation and expansion (2). Previous studies showed that c-Jun stimulates Nanog transcription through its binding to an AP-1-like site (M1) on the Nanog promoter and Nanog expression alone induces the clonogenic formation of colon cancer cells (16). Our recent studies also demonstrated that p38γ trans-activates MMP9 (43) and cyclin D1 (11) through c-Jun mediated binding to their respective promoters at the AP-1 site. Because p38γ stimulates c-Jun expression and interacts with Hsp90, Ras, and c-Jun (Figs.1A/4A and Supplementary Fig.S4C) (11), p38γ immunoprecipitates were analyzed for their complex formation in cells and at the Nanog promoter DNA. Moreover, one group of cells was treated with PFD to explore if PFD inhibits CSC expansion through regulating this complex binding to the Nanog promoter. Results (Fig.6A) show that the p38γ/Hsp90/Ras complex also contains c-Jun protein, which is disrupted by incubation with PFD, indicating that p38γ activity is required for this complex formation as demonstrated in colon tumors (22). Of great interest, this complex also binds the Nanog promoter at the AP-1 site (Fig.6B), indicating that p38γ may stimulate CSC expansion and TNBC progression through directly binding to the Nanog promoter. This conclusion is further supported by the fact that the PFD treatment decreases Nanog expression, CSC population, and TNBC progression (Figs.3G/H/5; Supplementary Figs.1/5/S6) and blocks p38γ as well as c-Jun bindings to the promoter (Fig.6B). Moreover, p38γ expression increases, while its depletion decreases the Nanog promoter luciferase activity, and the p38γ stimulatory effect becomes inhibitory when the AP-1 site on the promoter is mutated (Figs.6C/D and Supplementary Fig.S7C) (16). Together, these results indicate that p38γ may stimulate CSC expansion through trans-activation of Nanog through c-Jun/AP-1, which may play a critical role in TNBC development and progression (Fig.6E).

Figure 6. p38γ stimulates Nanog transcription through c-Jun/AP-1.

Figure 6

Open in a new tab

A, cells were incubated with PFD or DMSO for 24 h, which were subjected to p38γ immunoprecipitation and WB analysis. B, cells were treated as in A and then subjected to Chip assays with indicated antibodies. Precipitated DNAs were analyzed by PCR using primers covering the AP-1 like site as indicated (bottom). C, D, indicated cells were transiently transfected with the wild-type (WT) or the AP-1 site mutated (MT) human Nanog luciferase promoter together with p38γ. Luciferase activity was assessed 48 h later (mean ± SD, n = 3). E, an experimental model indicates that p38γ stimulates Nanog transcription through a multiple-protein complex and activated Nanog will then stimulate Sox2 and Oct3/4 expression through a transcriptional circuitry leading to CSC initiation and expansion resulting in TNBC development and progression. The p38γ inhibitor PFD will disrupt the complex, decrease Nanog expression, and inhibit CSC expansion and TNBC growth.

Discussion

TNBC lacks established targets for therapeutic intervention and consequently has the poorest prognosis among all types of breast cancer (5, 44). Here, our studies provide several pieces of evidence that together with published results indicate that p38γ may be a novel TNBC therapeutic target by stimulating CSC expansion through a mechanism involving trans-activating Nanog. First, p38γ collaborates with Ras to stimulate TNBC invasion (10) and metastasis (Supplementary Figs.S3B-D), and overexpressed p38γ in TNBC has been further demonstrated through systemic screenings and pathological analysis (11-14, 45). Further, p38γ is required both for malignant progression and CSC expansion in TNBC (Figs.1/3 and Supplementary Figs.S1/3). Moreover, p38γ forced-expression alone is sufficient to stimulate CSC expansion and to induce the malignant transformation of human mammary epithelial cells in vitro and/or in vivo (Fig.2 and Supplementary Fig.S2). In addition, p38γ requires its activity to stimulate CSC expansion, to promote TNBC progression, and to form a multiple-protein complex at the Nanog promoter through c-Jun/AP-1 (Figs.3-6). Furthermore, the p38γ specific pharmacological inhibitor PFD selectively inhibits TNBC growth in cell culture and/or in mice and reduces the CSC population in TNBC cells in association with disruption of the complex at the Nanog promoter (Figs.5/6 and Supplementary Figs.5/6/7A/B). Together, these results indicate that p38γ may promote TNBC development and progression by stimulating CSC expansion and may therefore be a novel therapeutic target for TNBC. Targeting CSC by the p38γ inhibitor PFD may consequently be a novel strategy for the treatment of TNBC (Fig.6E).

Our findings that p38γ promotes TNBC by stimulating CSC expansion through c-Jun-mediated binding to the Nanog promoter are fundamentally important for understanding the role of CSC in TNBC. CSC cells characterized by CD44high/CD24low are known to be abundant in TNBC (20, 46), which is further confirmed by using CD133 as a separate CSC marker (47). Additional studies showed that PKCα connects CSC to TNBC through an AP-1 dependent pathway (7) and that the stress-related protein XBP1 maintains the CSC population in TNBC as a co-regulator of hypoxia inducible factor 1α (48). These results together indicate a critical role of CSC in driving TNBC development and progression through a transcriptional program. However, the exact mechanisms by which this program links CSC and TNBC remain unknown. Here, we found that p38γ MAPK directly stimulates Nanog transcription to promote CSC expansion through c-Jun-mediated binding of a protein complex to the AP-1 site of the Nanog promoter (Fig.6). The functional role of this pathway is further demonstrated by a suppression of CSC expansion and TNBC growth/progression through p38γ depletion and/or inhibition with PFD in association with a blockade of p38γ/c-Jun binding to the Nanog promoter leading to decreased Nanog gene expression (Figs.1-6). However, roles of Hsp90 and Ras in this multiple-protein complex (Figs.6A/B) are not completely understood. Since Ras stimulates p38γ expression (8, 10) and elevated p38γ maintains Ras expression by activating Hsp90 (22), Hsp90 and Ras may cooperate to maintain elevated p38γ levels in the complex for its sustained activity to stimulate Nanog transcription as a key CSC driver. This notion is consistent with the involvement of Ras in CSC expansion (49, 50) and in TNBC progression (51), and with the TNBC growth-inhibitory activity of Hsp90 inhibitors (Fig.4C) (38). Further studies are warranted to investigate if Ras activates CSC expansion through its complex-formation with p38γ/c-Jun/Hsp90 on the Nanog promoter.

Our results that the p38γ specific pharmacological inhibitor PFD selectively inhibits TNBC-growth and significantly decreases CSC population in TNBC cells (Figs.1/5) have strong application potentials in clinic. PFD is orally active, relatively non-toxic, and is currently in phase III clinical trials as an anti-fibrotic agent (40, 42). Although PFD was previously shown to selectively inhibit p38γ activity in vitro (21), its cellar targets have not been identified (41). Our recent studies showed that PFD can efficiently inhibit the p38γ MAPK activity to phosphorylate several substrates, including ER (11), PTPH1 (9) and Hsp90 in cancer cells and/or in tumor xenografts (22). Most importantly, we found that PFD alone at non-toxic doses significantly blocks the growth of two TNBC xenografts in association with an inhibition of the CSC signature (Figs.5B-E and Supplementary Figs.S5A/B/S6). These results indicate that PFD may be repurposed to target CSC population by inhibiting p38γ activity in TNBC, which would have an immediate benefit for the treatment of TNBC in clinic. Moreover, Hsp90 inhibitors can selectively eradicate the stem cells (52) and effectively inhibit TNBC tumor-growth (38). We further showed that Hsp90 activity is required for endogenous and p38γ-induced Ras expression in MCF10A and TNBC cells through a complex formation (Figs.4-6 and Supplementary Fig.S4). These results further indicate that a combination of an Hsp90 inhibitor with the non-toxic p38γ inhibitor PFD may even be more effective for the treatment of TNBC through disrupting the functional protein-complex thereby synergistically depleting the CSC population (Fig.6E).

Conclusions

In conclusion, we have demonstrated that p38γ MAPK is a novel therapeutic target of TNBC in vitro and in mice. Moreover, we provide evidence indicating that p38γ may promote TNBC development and progression through a mechanism involving stimulating CSC expansion by increasing Nanog transcription. Targeting p38γ by its specific, non-toxic and FDA-approved pharmacological inhibitor PFD may be a new therapeutic strategy for the treatment of TNBC.

Supplementary Material

S1
S2
S3
S4
S5
S6
S7
SI

Acknowledgments

This work was supported by Award Number I01BX002883 from the Biomedical Laboratory Research & Development Service of the VA Office of Research and Development, and by grants from Cancer Center of Medical College of Wisconsin and Clinical & Translational Science Institute of Southeast Wisconsin (CTSI) (to GC). We would like to thank Drs. Jiahuai Han (Scripps Research Institute), Robert Weinberg (MIT), and Abdolrahman Nateri (University of Nottingham, UK) for providing reagents that made this work possible.

Footnotes

Authors’ contribution:

Xiaomei Qi: Conception and design, performing experiments, and data analysis

Ning Yin: Performing experiments and data analysis

Shao Ma: Performing experiments and data analysis

Adrienne Lepp: Performing experiments and data analysis

Jun Tang: Data analysis and interpretation

Weiqing Jing: Performing experiments and data analysis

Bryon Johnson: Data analysis and interpretation

Michael Dwinell: Data analysis and interpretation

Christopher Chitambar: Data analysis and interpretation

Guan Chen: Conception, design, data analysis and interpretation, manuscript writing, and manuscript writing

Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed.

References

  • 1.Ailles M, Weissman IL. Cancer stem cells in solid tumors. Curr Opin Biotechnol. 2007;18:460–6. doi: 10.1016/j.copbio.2007.10.007. [DOI] [PubMed] [Google Scholar]
  • 2.Young RA. Control of the embryonic stem cell state. Cell. 2011;144:940–54. doi: 10.1016/j.cell.2011.01.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Takebe N, Harris PJ, Warren RQ, Ivy SP. Targeting cancer stem cells by inhibiting Wnt, Notch, and Hedgehog pathways. Nat Rev Clin Oncol. 2011;8:97–106. doi: 10.1038/nrclinonc.2010.196. [DOI] [PubMed] [Google Scholar]
  • 4.Schieber MS, Chandel NS. ROS links glucose metabolism to breast cancer stem cell and EMT phenotype. Cancer Cell. 2013;23:265–7. doi: 10.1016/j.ccr.2013.02.021. [DOI] [PubMed] [Google Scholar]
  • 5.Foulkes WD, Smith IE, Reis-Filho JS. Triple-negative breast cancer. N Engl J Med. 2010;363:1938–48. doi: 10.1056/NEJMra1001389. [DOI] [PubMed] [Google Scholar]
  • 6.Lehmann BD, Bauer JA, Chen X, Sanders ME, Chakravarthy AB, Pietenpol JA. Identification of human triple-negative breast cancer subtypes and preclincal models for selection of targeted therapies. J Clin Invest. 2011;121:2750–67. doi: 10.1172/JCI45014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Tam WL, Lu H, Buikhuisen J, et al. Protein kinase C a is a central signaling node and therapeutic target for breast cancer stem cells. Cancer Cell. 2013;24:347–64. doi: 10.1016/j.ccr.2013.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Tang J, Qi X, Mercola D, Han J, Chen G. Essential role of p38γ in K-Ras transformation independent of phosphorylation. J Biol Chem. 2005;280:23910–7. doi: 10.1074/jbc.M500699200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Hou S, Padmanaban S, Qi X, Leep A, Mirza S, Chen G. p38γ MAPK signals through phosphorylating its phosphatase PTPH1 in regulating Ras oncogenesis and stress response. J Biol Chem. 2012;287:27895–905. doi: 10.1074/jbc.M111.335794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Qi X, Tang J, Loesch M, Pohl N, Alkan S, Chen G. p38γ MAPK integrates signaling cross-talk between Ras and estrogen receptor to increase breast cancer invasion. Cancer Res. 2006;66:7540–7. doi: 10.1158/0008-5472.CAN-05-4639. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Qi X, Zhi H, Lepp A, et al. p38γ mitogen-activated protein kinase (MAPK) confers breast cancer hormone sensitivity by switching estrogen receptor (ER) signaling from classical to nonclassical pathway via stimulating ER phosphorylation and c-Jun transcription. J Biol Chem. 2012;287:14681–91. doi: 10.1074/jbc.M112.349357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Meng F, Zhang H, Liu G, et al. p38γ mitogen-activated protein kinase contributes to oncogenic properties maintenance and resistance to poly (ADP-ribose)-polymerase-1 inhibition in breast cancer. Neoplasia. 2011;13:472–82. doi: 10.1593/neo.101748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Rosenthal DT, Lyer H, Escudero S, et al. p38γ promotes breast cancer motility and metastasis through regulation of RhoC GTPase, cytoskeletal architecture, and a novel leading edge behavior. Cancer Res. 2011;71:6338–49. doi: 10.1158/0008-5472.CAN-11-1291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Lee ST, Feng M, Wei Y, et al. Protein tyrosine phosphatase UBASH3B is overexpressed in triple-negative breast cancer and promotes invasion and metastasis. Proc Natl Asso Sci USA. 2013;110:11121–6. doi: 10.1073/pnas.1300873110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Li Z, Jiang Y, Ulevitch RJ, Han J. The primary structure of p38γ: a new member of p38 group of MAP kinases. Biochem Biophys Res Commun. 1996;228:334–40. doi: 10.1006/bbrc.1996.1662. [DOI] [PubMed] [Google Scholar]
  • 16.Ibrahim E, Babaei-Jadidi R, Saadeddin A, Spencer-Dene B, Hossaini Sea. Embryonic NANOG activity defines colorectal cancer stem cells and modulated through AP1- and TCF-dependent mechanisms. Stem Cells. 2012;30:2076–87. doi: 10.1002/stem.1182. [DOI] [PubMed] [Google Scholar]
  • 17.Hou SW, Zhi H, Pohl N, et al. PTPH1 dephosphorylates and cooperates with p38γ MAPK to increases Ras oncogenesis through PDZ-mediated interaction. Cancer Res. 2010;70:2901–10. doi: 10.1158/0008-5472.CAN-09-3229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Wolf J, Dewi DL, Fredebohm J, et al. A mammosphere formation RNAi screen reveals that ATG4A promotes a breast cancer stem-like phenotype. Breast Cancer Res. 2013;15:R109. doi: 10.1186/bcr3576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Liu Y, Chen C, Qian P, Lu X, Sun B, Zhang Xea. Gd-metallofullerenol nanomaterial as non-toxic breast cancer stem cell-specific inhibitor. Nat Comm. 2015;6:5988. doi: 10.1038/ncomms6988:1-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ibrahim S, Hassan H, Vilardo L, Kumar SK, Kumar AV, Reinbold Rea. Syndecan-1 (CD138) modulates triple-negative breast cancer stem cell properties via regulation of LRP-6 and IL-6-mediated Stat3 signaling. PloS ONE. 2013;8:e85737. doi: 10.1371/journal.pone.0085737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Ozes O, Blatt LM, Seiwert SD. Use of pirfenidone in therapeutic regimens. 7,407,973 B2 United States Patent-US. 2008 Aug.:1–46.
  • 22.Qi XM, Xie C, Hou S, et al. Identification of a ternary protein-complex as a therapeutic target for K-Ras-dependent colon cancer. Oncotarget. 2014;5:4269–82. doi: 10.18632/oncotarget.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Gough DJ, Corlett A, Schlessinger K, Wegrzyn J, Larner AC, Levy DE. Mitochonchrial START3 supports Ras-dependent oncogenic transformation. Science. 2009;342:1713–6. doi: 10.1126/science.1171721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhang J, Liang Q, Lei Y, Yao M, Li L, Gao Xea. SOX4 induces epithelial-mesenchymal transition and contributes to breast cancer progression. Cancer Res. 2012;72:4597–608. doi: 10.1158/0008-5472.CAN-12-1045. [DOI] [PubMed] [Google Scholar]
  • 25.Mani SA, Guo W, Liao M, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;704:704–15. doi: 10.1016/j.cell.2008.03.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Scheel C, Eaton EN, Li SH, et al. Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast. Cell. 2011;145:926–40. doi: 10.1016/j.cell.2011.04.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Guo W, Kechesova Z, Donaher JL, Shibue T, Tischler V, Weinberg RA. Slug and sox9 cooperatively determine the mammary stem cell state. Cell. 2012;148:1015–28. doi: 10.1016/j.cell.2012.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Shin S, Dimitri CA, Yoon S, Dowdle W, Blenis J. ERK2 but not ERK1 induces epithelial-to-mesenchymal transformation via DEF motif-dependent signaling events. Mol Cell. 2010;38:114–27. doi: 10.1016/j.molcel.2010.02.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Velasco-Velaque MA, Popov VM, Lisanti MP, Pestell RG. The role of breast cancer stem cells in metastasis and therapeutic implications. Am J Pathol. 2011;179:2–11. doi: 10.1016/j.ajpath.2011.03.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Gao H, Chakraborty G, Lee-Lim AP, Mo Q, Decker M, Vonica Aea. The BMP inhibitor Coco reactivates breast cancer cells at lung metastatic sites. Cell. 2012;150:764–79. doi: 10.1016/j.cell.2012.06.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.von Lintig FC, Dreilinger AD, Varki NM, Wallace AM, Casteel DE, Boss GR. Ras activation in human breast cancer. Breast Cancer Res Treat. 2000;62:51–62. doi: 10.1023/a:1006491619920. [DOI] [PubMed] [Google Scholar]
  • 32.Sivaraman VS, Wang H, Nuovo GJ, Malbon CC. Hyperexpression of mitogen-activated protein kinase in human breast cancer. J Clin Invest. 1997;99(7):1478–83. doi: 10.1172/JCI119309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Mueller H, Flury N, Eppenberger-Castori S, Kueng W, David F, Eppenberger U. Potential prognostic value of mitogen-activated protein kinase activity for disease-free survival of primary breast cancer patients. Int J Cancer. 2000;89:384–8. doi: 10.1002/1097-0215(20000720)89:4<384::aid-ijc11>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
  • 34.Morel A, Lievre M, Thomas C, Hinkal G, Ansieau S, Puisieux A. Generation of breast cancer stem cells through epithelial-mesenchymal transition. Plos One. 2008;3(8):e2888. doi: 10.1371/journal.pone.0002888. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Cordenonsi M, Zanconato F, Azzolin L, Forcato M, Rosato Aea, Picccolo S. The hippo transducer TAZ confers cancer stem cell-related traits on breast cancer cells. Cell. 2011;147:759–72. doi: 10.1016/j.cell.2011.09.048. [DOI] [PubMed] [Google Scholar]
  • 36.Zhang J, Liang Q, Lei Y, Yao M, Li L, Gao Xea. Sox4 induces epithelial-mesenchymal transition and contributes to breast cancer prevention. Cancer Res. 2012;72:4597–608. doi: 10.1158/0008-5472.CAN-12-1045. [DOI] [PubMed] [Google Scholar]
  • 37.Trepel J, Mollapour M, Giaccone G, Neckers L. Targeting the dynamic HSP90 complex in cancer. Nat Rev Cancer. 2010;10:537–49. doi: 10.1038/nrc2887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Caldas-Lopes E, Cerchietti L, Ahn JA, et al. Hsp90 inhibitor PU-H71, a multimodal inhibitor of malignancy, induces complete responses in triple-negative breast cancer models. Pro Natl Acad Sci USA. 2009;106:8368–73. doi: 10.1073/pnas.0903392106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Schulte TW, Blagosklonny MV, Ingui C, Neckers L. Disruption of the Raf-1-Hsp90 molecular complex results in destablization of Raf-1 and loss of Raf-1-Ras association. J Biol Chem. 1995;270:24585–8. doi: 10.1074/jbc.270.41.24585. [DOI] [PubMed] [Google Scholar]
  • 40.Schaefer CJ, Ruhrmund DW, Pan L, Selwert SD, Kossen K. Antifibrotic activities of pirfenidone in animal models. Eur Respir Rev. 2011;20:85–97. doi: 10.1183/09059180.00001111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Moran N. p38 kinase inhibitor approved for idiopathic pulmonary fibrosis. Nat Biotechnol. 2011;29:301. doi: 10.1038/nbt0411-301. [DOI] [PubMed] [Google Scholar]
  • 42.Noble PW, Albera C, Bradford W, et al. Pirfenidone in patients with idiopathic pulmonary fibrosis (CAPACITY): two randomised trials. Lancet. 2011;377:1760–9. doi: 10.1016/S0140-6736(11)60405-4. [DOI] [PubMed] [Google Scholar]
  • 43.Loesch M, Zhi H, Hou S, et al. p38γ MAPK cooperates with c-Jun in trans-activating matrix metalloproteinase 9. J Biol Chem. 2010;285:15149–58. doi: 10.1074/jbc.M110.105429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Crown J, O'Shaughnessy J, Gullo G. Emerging targeted therapies in triple-negative breast cancer. Ann Oncology. 2012;23:56–65. doi: 10.1093/annonc/mds196. [DOI] [PubMed] [Google Scholar]
  • 45.Zhang H, Meng F, Liu G, et al. Forkhead transcription factor Foxq1 promotes epithelial-mesenchymal transition and breast cancer metastasis. Cancer Res. 2011;71:1292–301. doi: 10.1158/0008-5472.CAN-10-2825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Idowu MO, Kmieciak M, Dumur C, Burton RS, Grimes MM, Powers CNea. CD44+/CD24−/low cancer stem/progenitor cells are more abundant in triple-negative invasive breast carcinoma phenotype and are associated with poor outcome. Human Pathol. 2012;43:364–73. doi: 10.1016/j.humpath.2011.05.005. [DOI] [PubMed] [Google Scholar]
  • 47.Liu TJ, Sun BC, Zhao XM, et al. CD133+ cells with cancer stem cell characteristics associates with vasculogenic minicry in triple-negative breast cancer. Oncogene. 2013;32:544–53. doi: 10.1038/onc.2012.85. [DOI] [PubMed] [Google Scholar]
  • 48.Chen X, Iliopoulos D, Zhang Q, Tang Q, Greenblatt MB, Hatziapostolou Mea. XBP1 promotes triple-negative breast cancer by controlling the HIF1α pathway. Nature. 2014;508:103–7. doi: 10.1038/nature13119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Goel HL, Pursell B, Chang C, et al. GLI1 regulates a novel neuropilin-2/α6β1 integrin based autocrine pathway that contributes to breast cancer initiation. EMBO Mol Med. 2013;5:488–508. doi: 10.1002/emmm.201202078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Seguin L, Kato S, Franovic A, Camargo MF, Lesperance J, Elliott KCea. An integrin β3-KRAS-RalB complex drives tumour stemness and resistance to EGFR inhibition. Nat Cell Biol. 2014;16:457–8. doi: 10.1038/ncb2953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Bild A, Parker JS, Gustafson A, Acharya CR, Hoadley KA, Perou CM. An integration of complementary strategies for gene-expression analysis to reveal novel therapeutic opportunities for breast cancer. Breast Cancer Res. 2009;11:R55. doi: 10.1186/bcr2344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Newman B, Liu Y, Lee H, Sun D, Wang Y. Hsp90 inhibitor 17-AAG selectively eradicates lymphoma stem cells. Cancer Res. 2012;72:4551–61. doi: 10.1158/0008-5472.CAN-11-3600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Zhi H, Hou S, Li R, et al. PTPH1 cooperates with vitamin D receptor to stimulate breast cancer growth through their mutual stabilization. Oncogene. 2011;30:1706–15. doi: 10.1038/onc.2010.543. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1
NIHMS756948-supplement-S1.pdf (4MB, pdf)
S2
NIHMS756948-supplement-S2.pdf (9.8MB, pdf)
S3
NIHMS756948-supplement-S3.pdf (11.5MB, pdf)
S4
NIHMS756948-supplement-S4.pdf (1.7MB, pdf)
S5
NIHMS756948-supplement-S5.pdf (4.4MB, pdf)
S6
NIHMS756948-supplement-S6.pdf (3.4MB, pdf)
S7
NIHMS756948-supplement-S7.pdf (830.5KB, pdf)
SI
NIHMS756948-supplement-SI.docx (42.7KB, docx)

RESOURCES